Immunoglobulin (Ig) isotype switching occurs by a looping-out and deletion process which is focused on switch regions located upstream of the constant regions, with the exception of Cdelta. Little is known regarding the mechanism of switch recombination (SR). We recently devised a plasmid-based transient transfection assay for SR. Using this assay we showed that there are distinct switching activities which mediate mu->gamma3, mu->alpha mu->epsilon and mu->gamma1 switching. In preliminary results reported here we show that the switch plasmid assay is AID dependent. This observation strengthens the conclusion that the switch plasmid assay faithfully recapitulates physiological SR. These studies suggest that there is isotype specific recognition of S regions by switching activities. However, essentially nothing is known regarding how switching activities identify S regions. Aim I is designed to test whether the results obtained with the switch substrates can be confirmed in vivo. We will exchange the endogenous Sgamma3 region with Sgamma1 sequence using targeted homologous recombination. This experiment will allow us to determine whether the absence of the Sgamma1 specific switching factor will lead to an inability of the Sgamma3 endogenous locus to undergo SR when S?1 sequence is present in the S region. Such an outcome would confirm the existence of isotype specific switching factors in a physiological setting. Aim II in this application is focused on defining the nature of molecular recognition of S regions. Using the plasmid assay we plan to manipulate the size and sequence of the S regions to delineate a minimal target for SR. We will also create synthetic switch regions which are mutated at specific residues and we will locate the position of double strand breaks in the S DNA. Taken together, these studies will allow us to map functional recombination motifs (FRMs) which are located in the tandem repeats. In the third Aim we will use the FRMs defined in Aim II and ask whether DNA binding proteins interact with these motifs and mediate SR using gel shift and competition binding analyses. If the FRMs overlap with previously described SNIP and SNAP recognition motifs then we will ask whether SNIP and/or SNAP are functionally involved in SR. We will also explore the involvement of the FRM binding proteins by modulating their expression levels and determining whether recombination is affected as a consequence.